Computer Rating System Prediction Results for College Football (NCAA IA)

20 Season Totals

System Pct. Correct Against Spread Absolute Error Bias Mean Square Error games suw sul atsw atsl
Stat Fox0.748600.5234512.50550.6604252.748712533179346315
Line (updated)0.7401712.1945-0.1131244.573712527185
Ashby AccuRatings0.740170.5052512.39460.3556248.113712527185337330
Harmon Forcast0.740170.5192312.45080.1250248.833712527185351325
Pigskin Index0.738760.5278212.5477-0.3397255.480712526186351314
Edward Kambour0.738760.5228612.56490.4035258.935712526186366334
Computer Adjusted Line0.738400.4898012.2110-0.1322244.678711525186120125
Dokter Entropy0.734550.4878713.18421.2923279.344712523189342359
Thompson pred0.734550.5150612.3094-0.2660246.890712523189359338
System Average0.733150.5028612.5044-0.4384251.114712522190352348
Hank Trexler0.733150.4873312.52670.1587254.136712522190327344
Line (opening)0.731740.5099812.3371-0.3174254.296712521191281270
Thompson SPRS0.728930.5157112.3348-0.0259248.319712519193361339
Atomic Football0.728930.5035812.7171-0.5547256.456712519193352347
Ed Bemiss0.728550.5057812.9370-1.1206271.848711518193350342
Dave Congrove0.727530.5157112.7879-0.9156260.736712518194361339
Thompson stats0.726120.5085712.7880-0.5980263.100712517195356344
Tom Benson0.725740.4928613.19420.2078282.875711516195345355
Gupta Power Ratings0.725620.5212313.61680.1679304.141441320121221203
Keeper0.724720.5064212.92800.5575270.918712516196355346
System Median0.723310.5050412.4885-0.3887251.599712515197351344
Bihl System0.721950.4888312.6441-0.2441261.381410296114197206
Laz Index0.721910.4935812.7716-0.9722264.546712514198346355
CPA Rankings0.721910.4707613.2357-0.2511278.423712514198330371
PerformanZ Ratings0.720510.5021413.4099-0.9380291.193712513199352349
Stephen Kerns0.720510.4757113.70660.9094298.377712513199333367
Sagarin Predictive0.719100.5057113.0006-0.6604273.375712512200354346
Moore Power Ratings0.719100.4949912.7941-0.7912262.429712512200346353
Imes Comprank0.717910.4928413.0031-0.3282273.932709509200344354
ARGH Power Ratings0.716290.4948012.93540.0386269.148712510202333340
Howell0.716290.4932113.0301-0.2800270.901712510202327336
Lee Burdorf0.716290.4877713.1949-0.7062280.601712510202339356
Linear Regression0.716290.4542912.9378-0.1685272.502356255101159191
Sagarin0.714890.4828612.9426-0.6904268.494712509203338362
Born Power Index0.714890.4878713.1968-0.4187277.137712509203342359
Super List0.714890.4878714.45960.3736334.375712509203342359
Beck Elo0.713480.4814313.4789-0.8706282.709712508204337363
NutShell Sports0.710670.5371413.1498-0.3965273.241712506206376324
TSR Slots0.707870.5135519.7393-1.0590786.405712504208360341
NutShell Retro0.707870.5235413.53350.0573287.570712504208367334
Warren Claassen0.706460.4850213.8103-1.1496299.995712503209340361
Anderson/Hester *0.706270.4698013.2646-0.5668274.91930321489140158
Brent Craig0.705060.5037113.1201-1.4347280.696712502210339334
Massey Consensus0.705060.5007113.1910-0.7335275.559712502210351350
Dunkel Index0.704750.4834313.54490.2301292.122674475199321343
Marsee0.702250.4830113.85391.1376304.335712500212327350
Sagarin Elo0.700840.4842913.9876-0.5891308.018712499213339361
CPA Retro0.699440.4928613.6243-0.1013292.535712498214345355
Bassett Model0.695900.4942413.1823-0.0504272.107707492215343351
Massey BCS *0.694440.4465413.3517-0.4423277.43432422599142176
Billingsly0.692420.4828613.4777-0.6652280.262712493219338362
Colley Rankings *0.690140.4670513.4215-0.3746282.412355245110163186
Covers.com0.688720.5330313.4444-1.5875295.206665458207347304
Wolfe *0.686470.4664413.3491-0.0573279.83530320895139159
The Sports Report0.681180.4935814.2865-1.1090329.505712485227346355
Martien Maas0.676980.4685013.79090.3771297.789517350167238270
Least Squares w/ HFA0.651690.4885715.26960.5832368.957356232124171179
Logistic Regression0.648880.4914313.9408-0.1244317.623356231125172178
TSR Pred0.646070.4857120.0556-1.1878822.795712460252340360
Sports Report Elo0.646070.4757114.9124-0.2994341.967712460252333367
Sportrends0.584270.4756417.4001-8.8754478.658712416296332366

* This system does not make predictions.  I make predictions for this
  system by translating it to a new scale that allows for making predictions.



Retrodictive records are found by taking the ratings from the current week
and applying them to the entire season to date.

The ideal system would be one that has the highest correct game decisions,
has the smallest mean error(deviation from the actual game result), and has
a bias of zero.

Mean Error = average[abs(prediction-actual)]

      Bias = agerage(prediction - actual)

      Std. = Standard Deviation of individual game biases